Return to the USDOJ/OIG Home Page |
The U.S. Trustee Program's Efforts to Prevent Bankruptcy Fraud and Abuse
Report No. 03-17
March 2003
Office of the Inspector General
FINDING 2: PERFORMANCE MEASUREMENT SYSTEM
Tracking and Reporting Requirements According to the UST Manual, the EOUST is required to maintain a database to track criminal fraud referrals made by the UST Program to U.S. Attorneys, the FBI, or other law enforcement authorities.28 At the end of each calendar quarter, the EOUST provides each UST with a report detailing the referrals made by the region based on data submitted. The referrals are divided into open and closed categories. Each region or district office tracks its referrals until the referral is closed through prosecution or a decision by the U.S. Attorney to decline, or not prosecute, the case. The UST region or office is also to submit an updated report to the EOUST each quarter providing information on new referrals and the status of all open referrals, paying particular attention to those that are more than one year old. According to data from the Criminal Referral Tracking System, the USTs made 7,564 fraud referrals to law enforcement from January 1988 to September 2001.29 The EOUST uses the tracking system primarily for performance measurement reporting and for internal management reports. Based on the audit results presented later in this finding, the UST Program's data system is materially unreliable. However, we are reporting information from the system as the only available data on fraud referrals, convictions and potential loss. When system data is considered in context with other available evidence, the conclusions in this report are valid. The 7,564 fraud referrals made between 1988, when tracking began, and September 2001 resulted in 1,038 indictments and 837 convictions, or an overall conviction rate of 11 percent of the referrals. In addition, the UST Program identified about $11 billion in potential losses.30 The following table shows by regional office the number of referrals, convictions, conviction rates, and potential losses. The table shows considerable variation in the number of referrals, convictions, and conviction rates. Los Angeles had the most referrals at 2,834 and convictions at 164 - although the region's conviction rate was under 6 percent - and Cleveland had the least referrals at 94 with none resulting in a conviction. Cedar Rapids had the highest conviction rate, nearly 34 percent, although it produced only 145 referrals. Fourteen of the 21 regions made referrals that resulted in a conviction rate exceeding the UST Program average. According to unverified data in the tracking system, Denver at a reported $10 billion accounted for 93 percent of potential losses due to fraud.31 However, as discussed in the upcoming section of this report dealing with the tracking system, 72 percent of cases referred for prosecution did not identify the amount of loss.
UST Program's Data on Referrals, Convictions,
and Potential Loss by Regions As of September 2001
Fraud Referrals to Law Enforcement The UST regional offices are responsible for making fraud referrals to appropriate law enforcement agencies when there is evidence of a crime. The UST regional offices receive allegations of criminal activity from many sources, including third parties (creditors, ex-spouses, or ex-business partners), trustees, judges, and law enforcement agencies. Regional offices prefer to receive allegations in writing but may also accept complaints made orally. Also, regional offices sometimes discover criminal activity based on their review of debtors' bankruptcy petitions, trustees' operations, and petition preparers. Regional offices conduct a preliminary investigation to determine whether an allegation of fraud appears valid. The extent of the investigation varies depending upon the type of potentially criminal activity involved. For example, concealment of assets in bankruptcy proceedings may involve less investigative effort than a complicated ponzi scheme32 to defraud investors. The UST regional offices use a number of tools to aid them in the investigation, such as reviewing on-line investigative public records, verifying information on petitions and schedules with outside agencies such as the Social Security Administration, reviewing court docket information, and photographing concealed physical assets. If a regional office determines that fraud or another crime may have been committed, it prepares a referral package and forwards the referral to the U.S. Attorney, FBI, or other law enforcement agencies, if needed, for investigation and potential prosecution. The level of detail in the referral package varies by type of case. Also, some regions may not refer cases that do not meet the U.S. Attorney's threshold for prosecution either in terms of dollar amounts or type of offense. For example, an AUST in one region stated that only "good cases" are referred, meaning cases that are likely to be accepted by the U.S. Attorney for prosecution. Other regions opt to refer all cases. Two regional offices apply a three-level priority for referrals. The two offices prepare detailed criminal referral packages for Levels I and II, which are considered high priority because the referrals may meet the U.S. Attorney's minimum prosecutorial guidelines or involve special circumstances such as trustee defalcation, misuse of funds by an attorney or other professional, or other serious violations. Such higher-priority cases are more likely to result in prosecution. The two offices prepare abbreviated packages or a simple form for Level III criminal referrals, which are considered low priority because the referrals do not meet the minimum prosecutorial guidelines or lack strong evidence. Once the packages have been prepared, the fraud referral data is entered into the region's system. UST Program officials told us that criminal fraud prosecutions have been hampered by a lack of investigative resources, primarily from the FBI. The officials stated that after about a 6-month moratorium on bankruptcy fraud investigations by the FBI in 2001, the UST Program persuaded the FBI to once again investigate fraud referrals. The FBI had agreed to resume fraud investigations, but the events of September 11, 2001 caused the FBI to divert resources to counterterrorism-related efforts. The officials stated that FBI resources for fraud investigations remain limited. Criminal Referral Tracking System The EOUST uses the quarterly reports submitted by the regions to update the Criminal Referral Tracking System. The regional offices submit the data on spreadsheets designed by the EOUST. The EOUST performs a limited review of the 21 spreadsheets to ensure that the data fields are properly formatted so the spreadsheets can be downloaded into the EOUST's tracking system. However, the EOUST does not verify that the data is accurate, complete, or consistent. An EOUST staff member stated that the regions are responsible for ensuring the integrity of the data as the regions maintain the supporting documentation. Our review of the Criminal Referral Tracking System determined that the system is materially unreliable as a result of missing, inconsistent, and inaccurate data and lack of standard data entry protocols. In addition, the tracking system does not include data on the UST Program's efforts to investigate bankruptcy fraud cases prior to referring the case to law enforcement. Although the tracking system provides the only consolidated data available on fraud referrals nationwide, the data in the UST Program's fraud referral tracking system is of limited use due to omissions and errors. The tracking system contains 33 data fields. Some data fields are more critical than others for UST Program management and performance measurement. The EOUST captures data such as the date of the criminal referral, type of allegation, type of chapter, subject of the referral, type of subject, amount of the potential loss, date of the indictment, date of the conviction, and the statute of limitations. A complete listing of the 33 data fields is included in Appendix 8. Our review identified missing and inconsistent data for 19 of the 33 data fields in the tracking system, including fields that are vital to managing the UST Program and measuring performance such as the type of fraud, the classification of person committing the alleged fraud, indictments and convictions, and the amount of loss due to fraud. Because of these weaknesses in the data, the tracking system is not as effective a management tool as it could be with more reliable data. Appendix 9 provides a complete listing of the data problems we found. Examples of the problems follow.
These problems occurred mainly because the EOUST did not establish protocols for entering the data into the tracking system and did not monitor the quality of the data submitted by the regional offices. The regional offices independently determine the type of information that should or should not be included in the data fields. Therefore, the referral data was inconsistent. Also, the regional offices did not make a distinction if information not included was missing or unknown. As a result of the weaknesses in the tracking system, the UST Program is hampered in its ability to follow up on referrals, to perform meaningful trend analyses of fraudulent activity to ensure that resources are focused on the most serious and prevalent fraud and abuse problems, and to accurately measure UST Program performance. UST Program officials have stated that no one within the bankruptcy community has a true sense of how pervasive fraud is within the bankruptcy system. The UST Program is in the best position to provide accurate data on the extent of detected bankruptcy fraud because most allegations are identified by or reported through the 95 UST Program offices. However, due to the limitations of its tracking system, the UST Program does not currently have complete and reliable data on the fraud identified in the bankruptcy system. UST Program officials are aware of the limitations of the tracking system and acknowledge the need for improvements. These officials stated that their goal is to ensure that all referrals are accurate, properly researched and supported, and lead to prosecution and conviction. The officials stated that in reviewing the data system they determined that some data fields are not useful for program management or performance measurement. Consequently the UST Program plans to revamp and streamline the data system. Further, the officials pointed out that they must rely on the U.S. Attorney's offices or the courts to provide key data on the results of the referrals. Subsequent to the completion of our audit fieldwork, the UST Program began taking a number of actions to improve its data system, including issuing protocols to the field offices and placing the system under the supervision of a Chief Information Officer. Regional Offices' Referral Data Each quarter the regional offices send data on their fraud referrals to the EOUST so that the data can be downloaded into the Criminal Referral Tracking System. To determine the accuracy of the data submitted by the regional offices, we judgmentally selected a sample of referrals made by the five regional offices from FY 1999 through 2001. As stated in Finding 1, our sample size totaled 302 referrals made by the five regional offices. Where possible, we reviewed the case file for each of the 302 referrals. The case files varied widely in size depending on the nature of the allegation and the complexity of the investigation. Our review of the supporting documentation for the 302 fraud referrals at the regional level found that in 102 of 302 cases (nearly 34 percent) at least one data field was inaccurately completed. We could not verify data for another 27 cases (about 9 percent) due to missing case files. The following table shows the result of our review by region. Accuracy of the Fraud Referral Data
FYs 1999 to 2001
The five regional offices entered erroneous data or failed to enter required data for 102 referral cases. For example, in a few cases the regions entered the debtor as the subject of the referral when in fact the supporting documentation showed that the subject was a petition preparer, an attorney, or the individual could not be identified. In another example, the regional office determined that a debtor illegally transferred and concealed estate property, which resulted in a $1.5 million loss. However, the region did not include in the tracking system the name of the subject, the amount of loss, or the type of fraud. Referral data was inaccurate primarily because regional offices did not review the data to ensure its accuracy prior to submission to the EOUST. In one regional office the tracking system is maintained on the region's shared computer drive. Each of the five field offices within the region is responsible for entering its own criminal referral data into the spreadsheet. No one at the regional level is assigned to review criminal referral data to ensure that it is accurate and consistent prior to submitting the data to the EOUST. Although the specific data elements may vary in importance, the fact that some elements are inaccurate or missing undermines the reliability of the UST Program's overall fraud referral tracking system as discussed previously in the section of this report on the national tracking system. The UST Program may not be able to determine the extent of known fraud in the bankruptcy system without ensuring a reliable tracking and management system; reviews of the referral data by regional offices would help ensure the data is accurate and complete. Data on Bankruptcy Fraud Investigations As discussed previously, the Code (28 USC 586, 18 USC 3057, and 11 USC 1106) requires UST Program personnel to refer allegations of bankruptcy fraud to law enforcement authorities. Depending upon the region, the investigations are either conducted by field office or regional office personnel. In one of the regional offices we audited, a separate unit (Fraud Unit) was established to centrally manage bankruptcy fraud. The unit was responsible for:
The usefulness of the Criminal Referral Tracking System is limited because it does not demonstrate the extent and success of the UST Program's direct efforts to detect and investigate bankruptcy fraud. The EOUST designed the system to track only those cases that have been referred to law enforcement authorities. The EOUST did not design the tracking system to capture data on investigations conducted by UST Program personnel, nor did it require that regional offices track this data. We found that four of the five regional offices we audited did not track the investigations they conducted unless a referral was made to law enforcement authorities. Also, these offices did not always maintain files for the cases that were investigated but not referred to law enforcement. The files were usually destroyed because the regional offices did not see a need for the information and wanted to conserve storage space. As a result, these offices could not document the extent of their efforts to investigate bankruptcy fraud. Tracking UST investigations and outcomes could aid the UST Program in allocating resources. The resources needed to conduct the investigations can be extensive depending on the complexity of the case. For example, a case involving a stolen social security number, which most regional offices have encountered, would typically take about two weeks to investigate. However, a complex ponzi scheme would require about six months. One UST stated that resources required to investigate such cases are not taken into account for budgeting and staffing needs for regional offices. The current budget model is based on the number of bankruptcy cases that are assigned to the regions. The UST Program was aware that it did not have an accurate measure of its efforts to investigate and detect bankruptcy fraud. In October 2001, the EOUST required that regional offices track the number of criminal investigations that were initiated, including those that did not result in a referral to law enforcement. The request was made as part of the EOUST's Significant Accomplishment Report35 for the period covering October 1, 2001 through March 30, 2002. The Significant Accomplishment Report identifies the UST Program's efforts in civil enforcement, criminal fraud referrals, and other significant legal issues. The report, issued in May 2002, showed that the UST Program had initiated 961 criminal investigations during the semiannual period and that 550 (57 percent) investigations resulted in a referral to law enforcement. Tracking the criminal investigations would be useful to the UST Program for measuring its efforts to investigate and detect fraud, and the Criminal Referral Tracking System could be modified to include both investigations and referrals to avoid duplication of effort and to help ensure data accuracy. In an effort to enhance its anti-fraud efforts, the UST Program's FY 2002 budget allocated $1 million to establish a Bankruptcy Fraud Team serving the 11 judicial circuits plus the Washington, D.C. area. The team is to consist of 12 attorneys who are responsible for identifying bankruptcy fraud and supporting the prosecutorial efforts of federal law enforcement authorities. According to the former Acting Deputy Director, one attorney position would be assigned to Washington, and the remaining 11 positions would be assigned to regional offices. He stated that a decision had not been made on allocating the 11 positions among the UST regions. As of May 2002, the positions had not been filled. A more comprehensive and accurate tracking system would have helped provide a basis for allocating these additional resources. Lastly, although the UST Program has begun to emphasize civil enforcement actions as an alternative or as a supplement to criminal fraud referrals, there is no nationwide system to track such actions or to compile performance data. Instead, the regions have developed their own civil enforcement tracking systems. Some regions have established multiple systems for the various types of civil enforcement actions taken. Use of the Data for Performance Measurement The UST Program prepared a Performance Plan for FY 2002 and FY 2003, and incorporated the plan into its budget submission. The UST Program established a performance goal to ensure that parties adhere to the standards of the law and to police for embezzlement, fraud, and other abuses. To meet the performance goal, the UST Program developed several performance indicators. The performance indicators for FY 2002 and FY 2003 are shown in the following table. FY 2002 and 2003
Performance Indicators
Notes: In accordance with the Department's overall policy, the UST Program did not establish targets for the number of criminal referrals and convictions but instead reported data on a prior-year basis. The Department issued a policy letter in February 1999 stating that numerical targets would not be established for performance indicators relating to law enforcement activities such as arrests, indictments, convictions, and seizures. The Department created this policy out of concern that such targets could be seen as "bounty hunting." However, the policy does allow components to collect and report on these law enforcement activities on a prior-year basis. The FY 2003 performance plan showed that in FY 2001, the UST Program made 586 referrals to law enforcement and 45 referrals resulted in a conviction. However, the referral data is extracted from the Criminal Referral Tracking System maintained by the EOUST, which we have found materially unreliable due to missing, inaccurate, and inconsistent data. As a result, Congress, the UST Program, and other decision-makers may rely on inaccurate and incomplete data to measure the UST Program's performance and to make funding decisions. Conclusions The data on fraud referrals and outcomes in the EOUST's Criminal Referral Tracking System is not fully reliable due to inaccurate, incomplete, and inconsistent data entry and a lack of monitoring by the EOUST and the regional offices. Further, the system lacks standard protocols for entering data, and the data is not adequately verified before submission. The weaknesses in the data system casts doubt on the ability of the UST Program and others to accurately measure the UST Program's performance. The system's limitations also affect the UST Program's ability to analyze trends in fraud and allocate resources accordingly. Recommendations We recommend the Director, Executive Office for U.S. Trustees:
Footnotes
|